Analysis of topics in health microeconomics through flexible regression models

Abstract

The object of discussion is To calculate a new flexible hospital production function by means of a Generalized Additive Model including interactions and to compare it with the classic models Cobb-Douglas and Translog in the prediction of the behavior of productive factors and to study how the number of beds in the hospital affects the hospital activity, the length of stays and, consequently, the waiting list.

The GAM model is more appropriate than the Cobb-Douglas or the translog to evaluate hospital production functions, for public hospitals located in Galicia and for the study period. The study also demonstrates the usefulness of simulation techniques to examine a hospital system.

There are no significant differences in terms of waiting lists and occupancy rates when the number of beds in larger hospitals increases. Supply-side policies can also be disappointing in their effects on waiting times for small rural hospitals. Greater capacity in terms of more beds is associated with shorter waiting times for medium size hospitals.